About SciencePG Frontiers
The column "SciencePG Frontiers" aims at presenting our authors’ latest publications of research achievements and speeding up the development of the science dissemination. SciencePG ensures that our authors get the recognition and rewards that they deserve and the opportunity to play a significant role in the global scientific community.
Publish your paper in SciencePG and write a promotional piece of news for your paper to gain more attention from the public.

Send your promotional news to: service@sciencepublishinggroup.com

Publication Services
Home / SciencePG Frontiers
Kernel Regression and Feed Forward Neural Networks Have Been Frequently Used in Modelling
Modelling methods have been important especially when non-parametric methods are applied because they are known to give better results than parametric methods and Body Mass Index models have been an increasing concern for health professionals.
By N. C. Mutono, Dr G. A. Waititu and Dr W .A. Kiberia
Jul. 16, 2016

Modelling methods have been important especially when non-parametric methods are applied because they are known to give better results than parametric methods and Body Mass Index models have been an increasing concern for health professionals.

In a recent paper by authors C. N. Mutono, G. A. Waititu and W. A. Kiberia, Body Mass Index was obtained by modelling using feed forward neural network and kernel regression methods which are non-parametric. This computation clearly shows the efficiency of using kernel regression methods over feed forward neural network.

“Kernel regression is better than feed forward neural network in modelling Body Mass Index and Body Mass Index modelling can be more accurate if people would consider the effect of other body dimensions other than height and weight alone because there are also other body dimensions that clearly show the level of body weight like waist size”, C. N. Mutono et. al said.

In the paper, C. N. Mutono et. al showed how Body Mass Index has been modelled using height and weight alone using both feed forward neural networks and kernel regression and they later show the effect adding twenty one variables which have a relationship with body weight. In modelling using kernel regression the model with more variables was the better model while using feed forward neural network the model with more variables was also a better model.

C. N. Mutono et. al goes on to suggest that kernel regression and feed forward neural network could effectively model Body Mass Index; However kernel regression was the better method.

Modelling of Body Mass Index is often seen as a first step of being able to determine the body fitness of an individual and this paper adds more value to Body Mass Index by being able to give more accurate statistical models.

Using non parametric methods like feed forward neural network and kernel regression helps in determining the best method to use when modelling. The advantage of using kernel regression lies in the ability of the method being able to get data driven bandwidth directly.

Mathematical modelling is playing an increasingly important role in helping health care personnel in understanding the effect of other body dimensions in calculating Body Mass Index. This will therefore bring rapid improvements in showing the body weights of people.

Gender has been a factor that determines the Body Mass Index of an individual and it has been found from the paper that the Body Mass Index of women is always higher that of men.

Authors:
N. C. Mutono, Student, Department of Statistics and Actuarial Science, Jomo Kenyatta University of Agriculture, kernel regression and feed forward neural network Research, Nairobi, Kenya.
Dr G. A. Waititu, Lecturer, Department of Statistics and Actuarial Science, Jomo Kenyatta University of Agriculture, kernel regression and feed forward neural network, Nairobi, Kenya.
Dr W .A. Kiberia, Lecturer, Department of Statistics and Actuarial Science, Jomo Kenyatta University of Agriculture, kernel regression and feed forward neural network, Nairobi, Kenya.

A paper about the study appeared recently in American Journal of Theoretical and Applied Statistics.

Paper link:
http://article.sciencepublishinggroup.com/html/10.11648.j.ajtas.20160504.13.html

ADDRESS
Science Publishing Group
548 FASHION AVENUE
NEW YORK, NY 10018
U.S.A.
Tel: (001)347-688-8931